Skip to main content

Full text of "The Montessori system examined"

See other formats


RETURN EDUCATION-PSYCHOLOGY LIBRARY 

TO— +• 2600 Tolman Hall 642-4209 



LOAN PERIOD 1 
SEMESTER 



SEMESTER 
m TF1 FPHON 



LCAN 



^RENEWALS 



ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 

2-hour books must be renewed in person 

Return to desk from which borrowed 

DUE AS STAMPED BELOW 



FACULTVlQANCt 



MAY 20 ^ 



SH0.IFCTT0RF 



FORM NO. DD10 



UNIVERSITY OF CALIFORNIA, BERKELEY 
BERKELEY, CA 94720 



1&ibzv$inz educational jttottoarap^ 

EDITED BY HENRY SUZZALLO 

PRESIDENT OF THE UNIVERSITY. OF WASHINGTON 
SEATTLE? W^tfHtNJCiTQJ* ! ; * *« 

THE 

MONTESSORI SYSTEM 

EXAMINED 



BY 

WILLIAM HEARD KILPATRICK, Ph.D. 

ASSOCIATE PROFESSOR OF THE PHILOSOPHY OF EDUCATION 
TEACHERS COLLEGE, COLUMBIA UNIVERSITY 




HOUGHTON MIFFLIN COMPANY 

BOSTON, NEW YORK AND CHICAGO 

<®bt fttoerswbe pte^ Cambri&0e 



&&*?, 






COPYRIGHT, I9I4, BY WILLIAM HEARD KILPATRICK 
ALL RIGHTS RESERVED 



Wbt 3Ribcrfiibe $re** 

CAMBRIDGE . MASSACHUSETTS 
U . S . A 



PREFACE 

The aim of this monograph is probably suffi- 
ciently indicated by the title. The purpose is 
to examine generally the educational doctrines 
promulgated by Dr. Maria Montessori, so as, 
first, to bring out their relation to one another 
and to other similar doctrines elsewhere held; 
and, second, to ascertain, as far as the author 
may, the contribution which Dr. Montessori has 
to offer to American education. 

My indebtedness, especially in proportion to 
the volume of matter, is great. To my colleagues 
on an investigating trip to Rome, Miss Annie E. 
Moore and Professor M. B. Hillegas, I am in- 
debted for very considerable assistance in the 
ordering of ideas and in reaching definite conclu- 
sions. My best thanks are due to the same two 
colleagues and to Professors John Dewey and 
Naomi Norsworthy for reading the manuscript 
and for making valuable suggestions. It would be 
unfair, however, to hold any one save the author 
responsible for the opinions herein expressed. 

W. H. K. 



544ti 



CONTENTS 

Editor's Introduction . vii 

I. Introduction i 

II. Education as Development .... 7^ 

III. The Doctrine of Liberty .... 12 

IV. Adequacy of Self-Expression in the 

Montessori System 27 

V. Auto-Education 31 

VI. Exercises of Practical Life ... 36 

VII. Sense-Training by Means of the Di- 
dactic Apparatus 42 

VIII. The School Arts: Reading, Writing, 

and Arithmetic 53 

IX. Conclusions 61 

Outline 68 



EDITOR'S INTRODUCTION 

The labors of Madam Montessori have aroused 
an unusual interest among Americans. Already 
her theories and practices are a frequent subject 
for investigation and discussion in meetings of 
teachers and parents. 

Among a considerable number of laymen and a 
smaller number of teachers, the interest amounts 
to enthusiasm. The doctrines of the Italian edu- 
cator are so warmly espoused by some that 
schools modeled on the plan of the Casa dei 
Bambini have been established in various parts 
of the country, where they rival and challenge 
the existing kindergartens and primary schools. 
To many of its adherents this movement consti- 
tutes an educational revolution which in time 
will completely change the education of children. 

The interest of the teaching profession as a 
whole is not marked by any such self-committal. 
The teachers are concerned to know the meaning 
of this agitation and are professionally curious 
to ascertain its worth for them. They are critical, 
if not skeptical; and they ask that the signifi- 
cance of this new expression of educational the- 
vii 



EDITOR'S INTRODUCTION 

ory be presented in terms of its practical bearing 
upon the teaching procedure commonly em- 
ployed with young children. They are tolerant 
enough of new dogma and experiment; but they 
possess a common-sense caution against a too- 
ready acceptance of them. They prefer to exam- 
ine a new program element by element, reserving 
the privilege of selecting and rejecting as their 
judgment decides. They would weigh every item 
of t the idealistic projects of radicals and even 
of the practical successes of experiments born 
among the differing conditions of foreign soil. 
Willing enough to admit that any new move- 
ment may contain factors that will aid in educa- 
tional evolution, they are not of the type com- 
pletely to let go of one institution in order to 
seize another. They prefer the safer position of 
being reconstructors of the old. 

While admitting the value of both types of 
thinkers and workers in the whole method of 
educational advance, it is to the relatively large 
group of public-school teachers and superintend- 
ents that this volume is addressed. 

The smaller class of heroic enthusiasts that 
become the more or less partisan leaders and fol- 
lowers of a new propaganda are not likely to be 
interested in a critical analysis of the particular 
viii 



EDITOR'S INTRODUCTION 

theories and practices that constitute their faith. 
With them the new institutional spirit is the 
thing! Details may be left to the rectification 
of time! 

Not so with the leaders and teachers of the 
rank and file! To them the detail is the thing! 
Upon the soundness of special theories and the 
effectiveness of particular practices, the strength 
of an institutional scheme depends. They want 
to know how far the theory of Madam Montessori 
departs from the best philosophy of education 
that the American profession knows. And when 
it does, they ask if experience, of both scientific 
and empirical sort, gives warrant to the varying 
belief. More than this, they would ascertain if 
claims made for practical success are proved; 
and, again, if such achievements may be repro- 
duced under the conditions of American life. 

These pertinent inquiries of American teachers 
require a judicial answer. It is offered in the 
brief accompanying volume, along with such 
historical and logical perspective as is necessary 
to clear understanding. 



THE MONTESSORI SYSTEM 
EXAMINED 

I 

INTRODUCTION 

The genesis of Madam Montessori's educational 
ideas is laid before the reader in simple but 
attractive manner in her principal work, The 
Montessori Method, as the English translation is 
called. But slight reference to the now well- 
known story is needed. Madam Montessori, as 
assistant physician at the Psychiatric Clinic of 
the University of Rome, became some fifteen 
years ago interested in defectives. She thus 
learned of the work done by Edward Seguin for 
the education of idiots. From this and from per- 
sonal experimentation in the education of feeble- 
minded^ there came the suggestion of using 
Seguin's method with the normal child. In this is 
found one important factor in the making of the 
Montessori method. While this study of defec- 
tives was going on, there had been organized in 
Milan a School of Scientific Pedagogy. The 
I 



THE MONTESSORI SYSTEM 

anthropologist Sergi appears to have been the 
leading spirit in the enterprise. The emphasis in 
this school was upon anthropometry and measure- 
ments in experimental psychology, particularly 
of the sensations. Whether from a more wide- 
spread interest or from the influence of this par- 
ticular school does not certainly appear, but the 
field of scientific measurement constitutes an- 
other factor in the formation of the Montessori 
method. A third element was the general back- 
ground of prevalent educational theory which 
one absorbs more or less unconsciously as one 
does his uncriticized religion or politics. This we 
may surmise was largely Pestalozzian in its ulti- 
mate origin. A fourth factor was the invitation 
extended to Madam Montessori by a building 
corporation in Rome to organize the infant 
schools in its model tenements. The effort to 
meet this demand created in large measure the 
Children's House, especially in its institutional 
aspect. In these four elements we seem to have 
the origin of the Montessori schools. 

It is not necessary to the purpose at hand to 
show just how far Madam Montessori is in- 
debted to Seguin for her didactic apparatus. No 
acknowledgment could be more open or generous 
than is hers; and every one acquainted with 
2 



INTRODUCTION 

Seguin's work will be struck with the similarity. 
There is, however, one important difference: 
Seguin was interested mainly in leading the 
defective to make those acquisitions of knowl- 
edge and skill which would with relative direct- 
ness prove useful in the ordinary affairs of his 
life; Madam Montessori, on the other hand, is 
more interested — as we shall later discover — 
in the disciplinary aspect of the exercises. 

The study of science has had far-reaching effect 
upon Madam Montessori and upon her educa- 
tional theory. In the general wish to apply 
scientific conceptions to education, few surpass 
her. Those who feel the urgent need for a more 
scientific study of education and for the bringing 
of the scientific spirit into our attitude toward 
educational practice, can but applaud the insist- 
ence with which Madam Montessori returns 
again and again to this point of view. In addition 
to the general demand for a scientific attitude on 
the part of teachers, we find specific elements of 
her procedure based on her scientific experience. 
For example, the teacher must keep records, both 
anthropometric and psychologic, of each child. 
The books in which these are kept are often 
shown to the visitor. The remark may be inter- 
jected that the data so recorded, unfortunately, 

3 



THE MONTESSORI SYSTEM 

hardly function otherwise than in keeping alive 
in the teacher a general spirit of child observa- 
tion. Another application of the scientific atti- 
tude is found in the insistence upon the liberty 
of the child as a prerequisite of the scientific study 
of educational data. "If a new and scientific 
pedagogy," says Madam Montessori, "is to 
arise from the study of the individual, such study 
must occupy itself with the observation of free 
children." Further, the adaptation of Seguin's 
material to a disciplinary end would seem to have 
had its origin in the wish on the part of Madam 
Montessori to utilize her scientific study of sense- 
experience. It must be said, however, that while 
Madam Montessori's interest in the scientific 
attitude is entirely praiseworthy, her actual 
science cannot be so highly commended. Her 
biology is not always above reproach, as, for 
example, the alleged disinfecting influence of 
garlic upon the intestines and lungs. She general- 
izes unscientifically as to the condition of con- 
temporary educational thought and practice 
from observation limited, it would seem, to the 
Italian schools. If she had known more of what 
was being thought and done elsewhere, her dis- 
cussions would have been saved some blemishes 
and her system some serious omissions. Her 

4 



INTRODUCTION 

psychology in particular would have been im- 
proved, had she known better what Wundt was 
doing in Germany, to mention no other names. 

While these shortcomings are mentioned, we 
should not fail to call attention to an evidence of 
scientific attitude and faith too seldom found in 
the teaching world — be it said to our shame. 
Few in the history of education have been capable 
of breaking so completely with the surrounding 
school tradition as has this Italian physician. To 
set aside tradition for science is no common 
achievement. That the innovator is a woman 
will seem to some all the more remarkable. With 
the true scientific spirit of experimentation 
Madam Montessori has devised a practice and 
an institution. Such a consciously scientific crea- 
tion stands in marked contrast with the conserva- 
tism and mystical obscurantism which but too 
widely characterize kindergarten education in 
America and elsewhere. Whatever opinion be 
held as to the success of the effort, no one can 
fail to approve Madam Montessori's thorough- 
going attempt to found a complete school 
procedure upon her highest scientific concep- 
tions. 

In the discussion which follows it will be as- 
sumed that the reader is acquainted with Madam 

5. 



THE MONTESSORI SYSTEM 

Montessori's chief work, The Montessori Method, 1 
and also with the didactic apparatus itself. The 
effort will be to examine the Montessori system 
and to appraise its worth to American education. 
Especial attention will be given to the merits of 
the Casa dei Bambini as a rival to the kinder- 
garten. Owing to limitations of space, only the 
most characteristic elements of the system will 
be considered. 

1 Frederick Stokes Company, New York, 191 2. 



II 

EDUCATION AS DEVELOPMENT 

That education should be considered as a devel- 
opment from within is a principal doctrine with 
Madam Montessori. The idea, of course, is an 
old one. Rousseau, Pestalozzi, and Froebel are 
among its most conspicuous exponents. The 
value of this point of view in the formation of our 
present educational practice is undoubted. The 
limitations of the doctrine, however, have not al- 
ways been clearly seen. Education as development 
has been likened to the care given to some rare 
and unknown plant. The gardener seeks to dis- 
cover and supply the conditions under which the 
plant can show its character or nature most com- 
pletely. But the analogy is clearly deficient, else 
anger and other ugly or erratic impulses should 
be expressed as completely and directly as those 
we prize more highly. The ill odor attaching to 
the word " whim" illustrates the point and shows 
the way. Life, indeed, consists in the expression 
of what we are, but under such conditions that 
the net result shall, in the long run, bring the 

7 



THE MONTESSORI SYSTEM 

fullest expression to all concerned. The condi- 
tions under which this proper expression may 
take place — so far as these have originated with 
man — make up the content of the cultural 
environment. Man has learned certain ways of 
doing things that he might the better express 
himself. This is as true of clothing, shelter, 
methods of procuring and preparing food, of art 
and literature, as it is of ethical concepts and 
legal procedure. The " funded capital of civiliza- 
tion' ' consists exactly of all the devices thus far 
contrived for the fullest expression of what we 
are, for our fullest possible development. 

Education is thus, in truth, the completest 
possible development of the individual; but the 
task of securing such a development is as great 
as is the complex of civilization. Expression 
involves as truly the mastering of this complex 
as it does the living-out of the impulsive life. 
More exactly, the two elements of mastering the 
environment and expressing one's self are but 
outer and inner aspects of one and the same proc- 
ess; each either meaningless or impossible apart 
from the other. Only in this larger sense can it 
be said that education is the development of the 
individual. 

Some, on the contrary, have taken the position, 
8 



EDUCATION AS DEVELOPMENT 

previously suggested, that in the child's nature 
as given at birth there is contained — in some 
unique sense — all that the child is to become, 
and this in such fashion that we should tend the 
child as the gardener does the plant, assured that 
the natural endowment would properly guide 
its own process of unfolding. Such is Madam 
Montessori's view. "The child is a body which 
grows and a soul which develops; ... we must 
neither mar nor stifle the mysterious powers which 
lie within these two forms of growth, but must 
await from them the manifestations which we 
know will succeed one another." "The educa- 
tional conception of this age must be solely that 
of aiding the psycho-physical development of the 
individual." "If any educational act is to be 
efficacious, it will be only that which tends to help 
toward the complete unfolding" of the child's 
individuality. 

Such a doctrine of education has borne good 
fruit; but there is danger in it. It has led in the 
past to unwise emphasis and to wrong practice. 
We have already seen that it carries with it a 
depreciation of the value rightly belonging to the 
solutions that man has devised for his ever- 
recurring problems. In fact, such a theory leads 
easily, if not inevitably, to Rousseau's opposition 

9 



THE MONTESSORI SYSTEM 

to man's whole institutional life. It further fail? 
to provide adequately for the most useful of mod 
ern conceptions, that of intelligent, self-directing 
adaptation to a novel environment. If develop- 
ment be but the unfolding of what was from the 
first enfolded, then the adaptation is made in 
advance of the situation, and consequently with- 
out reference to its novel aspects. Such a form 
of predetermined adaptation proves successful 
in the case of certain insects, as the wasp; for 
there the environment is relatively fixed. With 
man, however, each generation finds — and 
makes — a new situation. If education is to pre- 
pare for such a changing environment, its funda- 
mental concept must take essential cognizance 
of that fact. Still further, this erroneous notion 
of education gives to the doctrine of child liberty 
a wrong and misleading foundation. If the child 
already uniquely contains that which he is prop- 
erly destined to manifest, then the duty of the 
educator is to allow the fullest expression of 
what is implicitly given. But such a doctrine of 
liberty is notoriously disastrous. The result has, 
therefore, been that many have opposed every 
scheme of liberty in the schoolroom. By putting 
the demand for liberty on a false basis, its friends 
have too often proved its worst foes. It would not 
10 



\ 

EDUCATION AS DEVELOPMENT 

be fair to Madam Montessori to say that she 
herself draws all of these objectionable conclu- 
sions from her doctrine of the nature of education. 
She does not. She has not thought consecutively 
enough. But the conclusions are there to be 
drawn. They have been drawn from logically 
similar doctrines at other times. (We must, 
therefore, reject Madam Montessori's interpre- 
tation of the doctrine of development as inade- 
quate and misleading. The useful elements of this 
doctrine are covered up in error whenever devel- 
opment is identified with the mere unfolding of 
latency. 



Ill 

THE DOCTRINE OF LIBERTY 

The question here raised is that of the degree to 
which the child shall by his own choice determine 
his own activities at school. It was Rousseau 
who first brought this problem prominently for- 
ward. His advocacy of the educational utiliza- 
tion of liberty has profoundly influenced all sub- 
sequent thought. Froebel emphasized the same 
doctrine, but placed it rather on the false basis 
discussed in the preceding chapter. In contem- 
porary education, Professor Dewey is the most 
prominent exponent of the general point of view. 
In the preceding chapter we saw that some 
writers, including Madam Montessori, are in- 
clined to limit the concept of development to the 
mere unfolding of what has from the first been 
implicitly present. Such writers are further in- 
clined to consider the doctrine of liberty as 
simply a corollary to this conception of develop- 
ment. That is to say, if the child's whole future 
life is in fact already uniquely present in his 
nature at birth, then manifestly, that nature 

12 



THE DOCTRINE OF LIBERTY 

must be allowed to unfold. This is the point of 
departure for Madam Montessori's doctrine of 
liberty. "We cannot know," she says, "the 
consequences of suffocating a spontaneous action 
at the time when the child is just beginning to be 
active: perhaps we suffocate life itself. Humanity 
shows itself in all its intellectual splendor during 
this tender age . . . and we must respect reli- 
giously, reverently, these first indications of 
individuality. If any educational act is to be 
efficacious, it will be only that which tends to 
help toward the complete unfolding of this life. 
To be thus helpful it is necessary rigorously to 
avoid the arrest of spontaneous movements." 

But nearer, apparently, to Madam Montes- 
sori's heart is the liberty to be accorded the child 
as an object of scientific study. " The school 
must permit the free, natural manifestations of 
the child if in the school scientific pedagogy is 
to be born." The aim is to accord to the child 
" complete liberty." " This we must do if we are 
to draw from the observation of his spontaneous 
manifestations conclusions which shall lead to 
the establishment of a truly scientific child- 
study." A further reason for the use of liberty is 
found in Madam Montessori's belief that "disci- 
pline must come through liberty." By discipline 
13 



THE MONTESSORI SYSTEM 

she means self-control. "We call an individual 
disciplined when he is master of himself.' ' To 
this fruitful suggestion we shall later return. 

From these various considerations a system 
has been devised which accords a remarkable 
degree of freedom to the individual children of 
the Montessori schools. J A contrast between the 
Montessori school and the kindergarten of the 
more formal and traditional type may serve to 
give a clearer picture of the Montessori proced- 
ure, and consequently of the Montessori concep- 
tion of liberty as it appears in practice. The 
most evident difference is seen in the function 
of the teacher. The kindergartner is clearly the 
center and arbiter of the activity in the room. 
The Montessori directress seems, on the contrary, 
to be at one side. The kindergartner contem- 
plates at each moment the whole of her group; 
the directress is talking usually to one alone — 
possibly to two or three. The kindergarten chil- 
dren are engaged in some sort of directed group 
activity; each Montessori child is an isolated 
worker, though one or more comrades may look 
on and suggest. The arrangement of the room 
shows the same contrast. The kindergarten has 
a circle about which all may gather, and tables 
for group activity. The Montessori room is 

14 



THE DOCTRINE OF LIBERTY 

fitted, preferably, with individual tables, ar- 
ranged as the children will. (In the writer's 
observation, there has been little deviation, how- 
ever, from arrangement in formal rows.) Mon- 
tessori provides long periods, say of two or more 
hours, while the kindergarten period rarely goes 
beyond a half-hour. During the period assigned 
for that purpose practically all of the Montessori 
apparatus is available for any child (except for 
the very youngest or the newest comers), and the 
child makes his choice freely. The kindergartner, 
on the other hand, decides very nicely what 
specific apparatus shall be used during any one 
period. The Montessori child abides by his 
choice as long as he wishes, and changes as often 
as he likes; he may even do nothing if he prefers. 
The child in the traditional kindergarten uses the 
same apparatus throughout the period, and is 
frequently led or directed by the teacher as to 
what he shall do. At other times he may be at 
liberty to build or represent at will whatever may 
be suggested by the "gift" set for the period. 
The Montessori child, each at his own chosen 
task, works, as stated, in relative isolation, his 
nearest neighbors possibly looking on. The 
directress, perchance, will not interpose in the 
slightest throughout a whole period. In the kin- 

15 



THE MONTESSORI SYSTEM 

dergarten all the children at the table, for exam- 
ple, are directed — in the large, at least — by the 
teacher, and all keep more or less together in 
what they are doing. .The Montessori child learns 
self-reliance by free choice in relative isolation 
from the directress.'} He learns in an individual- 
istic fashion to respect the rights of his neighbors. 
The kindergarten child learns conformity to 
social standards mainly through social pressure 
focused and brought to bear in a kindly spirit by 
the kindergartner. His self-reliance tends to be 
the ease of mind resulting from conscious mas- 
tery of customs, adult-made and adult-directed. 
Consciousness of superiority, too, has at times its 
part in his self-reliance. It is thus clearly evident 
that in the Montessori school the individual child 
has unusually free rein. 

With so much liberty in the Montessori school 
it would be easy to suppose that anarchy must 
ensue. Such has not proved the case. In the 
first place, the directress is not to allow " useless 
or dangerous acts, for these must be suppressed, 
destroyed ." " The liberty of the child should have 
as its limit the collective interest." "Absolute 
rigor" is in extreme cases permitted. While 
these statements might be so interpreted as to 
imply coercion and suppression, there is in prac- 
16 



THE DOCTRINE OF LIBERTY 

tice little need of positive restraint, much less 
than one would have supposed. On a certain 
visit the writer saw one boy in a sudden temper 
pull another's hair, but the encounter subsided 
as quickly as it arose, and no notice was taken of 
the episode by any one. On the whole, the chil- 
dren worked as busily as ants about a hill. At 
times the noise would prove a little trying to one 
brought up in the belief that children should be 
seen and not heard. Probably, however, any 
protest against the noise would be rather conven- 
tional than just. To the writer the suggestion of 
great individual liberty proves very attractive. 

What is the desirable and feasible thing to do 
in this matter of liberty in school activities? It 
will perhaps suffice for our purpose to consider 
briefly four questions suggested by experience 
with the kindergarten: (i) Why allow the child 
to exercise his choice? (2) With free choice 
granted, how is cooperation in group activity to 
be secured? (3) How is the child to secure the 
requisite knowledge and skill? (4) How shall we 
secure conduct that conforms to social stand- 
ards? In strictness these questions seem hope- 
lessly to overlap; it is hoped, however, that this 
difficulty may be avoided in the discussion. 

Why allow the child to exercise free choice? It 

17 



THE MONTESSORI SYSTEM 

might be replied that the presumption of liberty 
lies with the individual; that any infringement 
must be justified. The writer would be willing 
to accept this reasoning, even in the case of the 
child. We need not, however, base our argument 
on this point of view. Other, and to some more 
convincing, considerations may be urged. In a 
democracy, self-direction must be the goal of 
education. How shall the child become self- 
directing? Can one learn to swim out of water? 
To become self-directing one must enter life 
itself, where decision and choice and responsibil- 
ity hold sway. This seems undoubted in the 
realm of conduct as an exercise of "will"; it is 
equally true of the more intellectualistic aspect 
of thinking. Under Professor Dewey's influence 
it has become a commonplace that no thinking 
worthy the name goes on apart from a felt 
problem, a thwarted impulse. The problems set 
by the teacher are too often not so felt by the 
children. A reported or artificial problem has 
little gripping effect. The real problem arises 
when the current of real life is for the time 
dammed. Under such conditions, the child puts 
heart and soul into the situation in a genuine 
effort to straighten things out. It is then, if ever, 
that there is training of "mind" or "will." But 
18 



THE DOCTRINE OF LIBERTY 

evidently the current of real life — in the sense 
here used — can flow only when the child has 
freedom to choose, to express himself. And life 
does not flow in twenty-minute periods. Let the 
child get genuinely interested, and the short 
period proves all too short. 1 If school life is to 
repeat and make possible actual life, the tyranny 
and artificiality of the short period and of over- 
much direction by the teacher must go. Real 
thinking and real conduct demand freer rein. 
Postponing for the time the discussion of Madam 
Montessori's curriculum, it appears that she and 
our more liberal American kindergartens are 
here well in advance of Froebel and the tradi- 
tional kindergarten. The absence of a detailed 
program and of excessive direction from above 
afford — in this respect, at least — a fuller oppor- 
tunity for genuine self-expression. 

The discussion so far given prepares for our 
second question: How shall we secure coopera- 
tion if the children be allowed freedom of choice? 
We now feel like turning the question about: 
How can cooperation be secured except by the 
spontaneous impulse of the children themselves? 

1 The objection here urged against the short period is based, 
of course, on an assumed regime of relative freedom. If tasks, 
however, are to be set, as is common in our schools, the short 
period may be a psychologic necessity. 

19 



THE MONTESSORI SYSTEM 

If cooperation be forced from without, is it not 
largely a sham and a counterfeit? The desirable 
group work is that joint activity which springs 
from the felt necessity of joint action. We are 
here but repeating the discussion of the preced- 
ing paragraph. What we wish, then, is to put the 
children into such socially conditioned environ- 
ment that they will of themselves spontaneously 
unite into larger or smaller groups to work out 
their life-impulses as these exist on the childish 
plane. From these considerations we criticize 
both Montessori and Froebel, the one that she 
does not provide situations for more adequate 
social cooperation, the other that the coopera- 
tion comes too largely from outside suggestion 
and from adult considerations. 

If our discussion of freedom has so far led us 
to emphasize the advantage of free choice, the 
two remaining of the four questions imply limita- 
tions in the exercise of such freedom. It has 
always been known that following one's own 
sweet will does not of necessity bring either the 
most of knowledge or the best of conduct. It 
is, indeed, the insistent obtrusion of this easily 
observed fact that has led parents and teachers 
in all times to set such severe limitations upon 
the free expression of the child's spontaneous 
20 



THE DOCTRINE OF LIBERTY 

impulses. If we were here concerned with the 
education of children of all ages, our task would 
be more difficult. As, however, we are more 
interested in the kindergarten age, the problem 
may not prove insoluble. 

Before asking how the child shall secure the 
requisite knowledge and skill, let us ask how 
much of these he should possess at the end of the 
sixth year? Is this so great in amount, or so 
difficult of acquisition, that only formal teaching, 
enforced by external compulsion, will suffice to 
give it? A child entering the primary school 
should — by common consent, at least — not be 
required to present very specific entrance prepa- 
ration. It is still true that he should have organ- 
ized and hold available a general range of experi- 
ence. We need not ask precise agreement, but 
in general he should have a certain use of the 
mother tongue. He should know the names and 
uses of many common things of ordinary life about 
him. He should know certain of the commonest 
physical properties of things. In certain ordinary 
manual activities it were well for him to have 
reasonable skill, using scissors, paste, a pencil or 
crayon, and colors. If he is able to stand in line, 
march in step, and skip, so much the better. He 
should know some enjoyable games and songs,, 

21 



THE MONTESSORI SYSTEM 

and some of the popular stories, suited to his age. 
He should be able, within reason, to wait on 
himself in the matter of bathing, dressing, etc. 
Propriety of conduct of an elementary sort is 
expected. 

Does any one question that knowledge and 
skill such as this can be gained incidentally in 
play by any healthy child? Indeed, so satisfied 
have many parents been of this point that they 
believe a kindergarten course unnecessary, feel- 
ing that home life suffices. Without accepting 
such a position, we may ask whether a group of 
normal children playing freely with a few well- 
chosen toys under the watchful eye of a wise and 
sympathetic young woman would not only ac- 
quire all this knowledge and skill and more, 
but at the same time be enjoying themselves 
hugely? Surely, to ask the question is to answer 
it. In this instance the doctrine of freedom is 
practically the doctrine of interest. As difficult 
as the problem of interest for the upper grades 
seems to be, here in the kindergarten age there is 
little difficulty apart from our lack of faith to try, 
or of skill to execute. We can leave a great deal 
more to the natural working-out of the child's 
spontaneous interest than many of us have dared 
to believe. And curiously enough, the kinder- 
22 



THE DOCTRINE OF LIBERTY 

gartner, in spite of FroebePs faith in childhood, 
is too often the opponent of real freedom. It is 
here as much as anywhere else that Madam 
Montessori, exemplifying Professor Dewey's 
teaching, will affect the practice of education in 
America. 

There yet remains the fourth question, the re- 
lation of free choice to right conduct. That the 
demands of propriety limit the natural freedom 
of conduct need not be questioned. The real 
question is, How can we so condition the child 
that he shall best be brought to observe the 
obligations that devolve upon him? In particu- 
lar, what is the relation of this proposed manage- 
ment to the child's spontaneity? The preceding 
considerations have disposed us to favor a rela- 
tively free expression of the childish nature. Is it 
different here? Shall we agree with the Director, 
Signor Stratico of Rome, in opposing the Mon- 
tessori system " because it makes little anarch- 
ists"? Perhaps democratic America had already, 
before the advent of Madam Montessori, arrived 
at a more approving attitude. 

We may as well admit at the outset that cer- 
tain of the child's natural impulses, probably 
acquired by the race under widely differing con- 
ditions of survival, cannot now be expressed in 

23 



fac 



idtp 



THE MONTESSORI SYSTEM 

the manner and at the time in which they nor- 
mally present themselves. Such manifestations 
we must either starve off, for the time suppress, 
or greatly redirect. Certain other impulses will 
need less of redirection; still others, only oppor- 
tunity for expression. Probably in the effort to 
suppress or redirect impulses a certain amount 
of positive pain association (" punishment") will 
prove necessary, particularly during the pre- 
kindergarten age; but, on the whole, the most 
effective plan of managing the recalcitrant im- 
pulses will be to encourage and feed those others 
which are naturally leading in the proper direc- 
tions. Thus again we find approval for positive 
self-expression. It is this principle that Madam 
Montessori has in mind when she speaks of 
" active discipline." Our fathers expressed the 
same in more theological guise when, speaking 
through Dr. Watts, they said, — 

"Satan finds some mischief still for idle hands to do." 

There is yet another and perhaps more impor- 
tant respect in which the principle of free expres- 
sion leads to proper habits of conduct. Ethics, 
propriety of conduct in general, is perhaps best 
conceived as the proper way of " getting along " 
with others, of adjusting one's self satisfactorily 
24 



THE DOCTRINE OF LIBERTY 

— to all concerned — to a social situation. If this 
be so, we could say almost a priori that only by 
mingling with people under normal conditions 
can one learn to "get on " with them. The stimuli 
of social approval and disapproval are, after all, 
about the strongest spurs for directing conduct 
aright that we know. Child and adult, alike, 
yield to the demand of their fellows. What we 
wish, then, is to put children of the kindergarten 
age under such conditions of companionship that 
they will learn gradually the fine art of living with 
their fellows. To this end adult supervision, true 
enough, will be necessary. It is just here that 
the kindergarten finds its chief raison d'etre. The 
teacher must at times intervene to draw dis- 
tinctions and direct wisely the course of ap- 
proval. The real agency, however, is the child's 
own comrades. 

It is difficult, then, to escape the conclusion, 
from whatever standpoint we view the situation, 
that the relatively free expression of the child's 
natural impulses — safeguarded, as discussed — 
is the efficient plan for his proper rearing. Such 
freedom is necessary if the child is to enter with 
full zest into actual cooperation, and into the 
acquisition of those habits of knowledge and 
skill which are properly to be expected. The same 

25 



THE MONTESSORI SYSTEM 

freedom is necessary if he is to grow into ade- 
quate self-reliance, and, at the same time, into the 
adequate control of self in the appreciation of the 
rights of others. From such considerations we 
highly approve Madam Montessori's reemphasis 
of the doctrine of liberty. In the practical out- 
working of her idea she has set an example to 
home, to kindergarten, and to primary school. 
There must be less of doing for the child where he 
can do for himself; less of the short-period pro- 
gram, where interest is too highly excited only 
to be too soon dissipated; less of minute direc- 
tion by mother, kindergartner, or teacher; — in 
short, more of opportunity for the child to lead a 
simple, healthy, normal life. 



IV 

ADEQUACY OF SELF-EXPRESSION IN THE 
MONTESSORI SYSTEM 

Freedom apart from self-expression is a contra- 
diction of terms. The discussion of Madam 
Montessori's doctrine of freedom given in the 
preceding chapter is, therefore, incomplete with- 
out a consideration of the adequacy of self- 
expression allowed by her system. The didactic 
apparatus which forms the principal means of 
activity in the Montessori school affords singu- 
larly little variety. Without discussing here the 
grounds for this restriction, it suffices to say that 
this apparatus by its very theory presents a 
limited series of exactly distinct and very precise 
activities, formal in character and very remote 
from social interests and connections. So narrow 
and limited a range of activity cannot go far in 
satisfying the normal child. It is, of course, true 
that a child finds pleasurable content in an activ- 
ity which to the adult would seem hopelessly 
formal, even to the point of drudgery. The small 
child who took off the box- top and put it back on 
27 



THE MONTESSORI SYSTEM 

for seventy-nine times in succession furnishes a 
good illustration. In the same way, we must not 
hastily conclude that no child could enjoy the rel- 
atively formal exercises of the didactic apparatus. 
Mechanical manipulation has strong attractions 
for childhood. But after all is said, the Mon- 
tessori school apparatus affords but meager diet 
for normally active children. Further, while 
happy childhood knows no stronger or more 
fruitful impulse than imaginative and construc- 
tive play, still, in these schools playing with the 
didactic apparatus is strictly forbidden, and usu- 
ally no other play-material is furnished. Madam 
Montessori has, in fact, been publicly quoted as 
saying, "If I were persuaded that children needed 
to play, I would provide the proper apparatus; 
but I am not so persuaded." The best current 
thought and practice in America would make 
constructive and imitative play, socially condi- 
tioned, the foundation and principal constituent 
of the program for children of the kindergarten 
age, but Madam Montessori rejects it. Closely 
allied with play is the use of games. One finds 
more attention paid to this, but the games seen 
in the Montessori schools of Rome are far inferior 
in every respect to those found in the better 
American kindergartens. Madam Montessori 
28 



ADEQUACY OF SELF-EXPRESSION 

herself seems, from her use of the very word 
"game," to have a most narrow and restricted 
conception of what games are, and of what they 
can do. Those more advanced forms of self- 
expression, drawing and modeling, are, on the 
whole, inferior to what we have in this country. 
Modeling is, in fact, hardly at all in evidence. 
Drawing and painting are occasionally good, but 
frequently amount to nothing but the coloring 
of conventionalized drawings furnished by the 
teacher. Stories have little or no place — a most 
serious oversight. There is very little of dramati- 
zationAOn the whole, the imagination, whether 
of constructive play or of the more aesthetic sort, 
is but little utilized. It is thus a long list of most 
serious omissions that we have to note. 

A partial offset to these deficiencies is found in 
the "practical life" activities. These undoubt- 
edly offer expression to a side of child nature too 
often left unsatisfied. To do something that 
counts in real life, not simply in the play world, 
is frequently one of the keenest pleasures to a 
child. It should be remarked, too, that the defi- 
ciencies noted are not inherent in the use of the 
more admirable features of the Montessori sys- 
tem; that is to say, we can borrow the good from 
this source without giving up the good we already 
29 



THE MONTESSORI SYSTEM 

have. Doubtless if Madam Montessori had her- 
self known more of better educational practice 
elsewhere, she would have incorporated some, 
perhaps all, of the features the absence of which 
we here regret. 

It is evident from the foregoing that, after all 
has been said, the Montessori curriculum affords 
very inadequate expression to a large portion of 
child nature | Such a limitation of opportunity is, 
in effect, nothing less than repression, a repres- 
sion destructive alike of happiness and mental 
growth. Moreover, since expression is the means 
to the acquisition of the culture of the race, the 
deficiency in expression is serious, whether it be 
looked at from the point of view of the child and 
his present happiness and growth, or from the 
point of view of culture and of the child's prepara- 
tion for participation therein. From every con- 
sideration, the proposed curriculum proves inad- 
equate and unduly restrictive. 



V 

AUTO-EDUCATION 

Auto - education as conceived by Madam 
Montessori is the necessary correlative of a r6- 
gime of freedom. From directed activity alone 
can training come, but for her direction must not 
contravene the child's freedom. With the teacher 
thus ruled out, and the child's self - direction 
inadequate, resort is had to the apparatus. In 
place of the old-time teacher, says Madam 
Montessori, "we have substituted the didactic 
material, which contains within itself the control 
of error, and which makes auto-education possi- 
ble to each child." 1 Does the reader ask how 
this is done? Let the cylinder box answer. This 
is a wooden block, in which are holes of varying 
depths. To each hole belongs a cylinder which 
exactly fills it. All the cylinders are removed, 
and the child proceeds to replace them. "If he 
mistakes, placing one of the objects in an opening 
that is too small for it, he takes it away, and pro- 
ceeds to make trial, seeking the proper opening. 
1 The Montessori Method, p. 371. 

31 



THE MONTESSORI SYSTEM 

If he makes a contrary error, letting the cylinder 
fall into an opening that is a little too large for it, 
and then collects all the successive cylinders in 
openings just a little too large, he will find him- 
self at the last with the big cylinder in his hand, 
while only the smallest opening is empty. The 
didactic material controls every error. The child 
proceeds to correct himself." 

The auto-education is for Madam Montessori 
the only true education. "This self-correction 
leads the child to concentrate his attention upon 
the differences of dimensions, and to compare the 
various pieces. It is in just this comparison that 
the psycho-sensory exercise lies." " It is the work 
of the child, the auto-correction, the auto-educa- 
tion which acts." 

It is impossible not to sympathize with Madam 
Montessori's intention in emphasizing this 
notion of auto-education. The more fully the 
child can learn from his own experience without 
any telling from the teacher, the more fully is 
his knowledge his own. If he can feel for himself 
the problem, if he can work out for himself a 
plan of solution, and if finally he can ascertain 
by tests of his own that his solution is correct — 
if these results can be attained from any plan, 
then surely that plan is a good one. 
32 



AUTO-EDUCATION 

When, however, we turn from the general con- 
ception to the specific working of Madam Mon- 
tessori's auto - education, we find that what 
should be the counterpart of generous liberty, 
amid many and varied opportunities, has in 
effect shrunk to a relatively mechanical manipu- 
lation of very formal apparatus. The didactic 
apparatus is in intention so devised that with 
each piece one, and only one, line of activity is 
feasible. Thus to the properly initiated child the 
sight of the cylinder box described above sug- 
gests only the taking-out of the cylinders and 
putting them back (any side suggestion, as 
improvising a wagon, is effectually suppressed). 
And the box is further so contrived that there is 
only one order in which the cylinders will fit. 
"The didactic material," in this case, at least, 
"controls every error." \It is in this limited 
fashion that Madam Montessori provides self- 
education. It is under such conditions that the 
directress keeps herself in the background and 
relies upon the cylinder box to set the problem 
and test the solution. Surely it is a naive trust in 
a very generous transfer of training which can 
see appreciable profit in so formal and restricted 
a scheme of auto - education. As applied by 
Madam Montessori, we must conclude, then, 

33 



THE MONTESSORI SYSTEM 

that auto-education is more of a wish than a fact. 
In her scheme it is too intimately bound up with 
the manipulation of the didactic apparatus to 
afford outside thereof the fruitful suggestion of 
wise procedure. 

If, on the other hand, we consider life itself 
and the situations that arise therefrom, we find 
abundant instances of evident self-education. A 
boy trying to drive a nail soon learns whether 
he is hitting it on the head. A pair of roller 
skates suggest their own problem with a mini- 
mum of explanation; they also test admirably 
the solution proffered. The sight of possible 
playmates suggests socially conditioned activity; 
the same children pass upon the newcomer's 
ability to participate successfully. We may 
generalize by saying that self-education is the 
concomitant of attempted purpose: whenever one 
can see the connection between effort and success 
he is on the road to the perfected activity. We 
are then led again to practically the same con- 
clusion as previously reached elsewhere: The 
nearer to the conditions of normal life that the 
school life can be brought, the more will real 
problems present themselves naturally (and not 
artificially at the say-so of the teacher). At the 
same time, the practical situation which sets the 

34 



AUTO-EDUCATION 

problem will test the child's proposed solution. 
This is life's auto-education and a right good 
pedagogic scheme it is. If Madam Montessori's 
term and general discussion can help us attain 
in practice what we have for years admitted in 
theory, she will have an honorable part in the 
reorganization now under way of our kinder- 
garten and early primary education; but the 
formal auto-education based on the didactic 
apparatus is at present more of a danger than 
a help. 



VI 

EXERCISES OF PRACTICAL LIFE 

The Montessori schools were first devised in con- 
nection with an unusually intelligent effort at 
improving certain tenement houses in Rome. 
The families being poor, it was an assistance to 
them for the school to take care of the children 
during as much of the day as possible. Accord- 
ingly, the length of the school day advocated 
by Madam Montessori extended to eight or ten 
hours according to season. There was thus a 
concentration of authority and responsibility in 
the school, which was the more fortunate, since 
many of the parents had low standards of living. 
In this way, the " Children's Houses" pay much 
attention to cleanliness of person and dress. The 
children are taught to wash their hands, brush 
their hair, brush their teeth, rinse their mouths, 
and otherwise care for their bodily and personal 
needs. The schoolroom is largely kept in order 
by the children themselves. Since school is 
held during practically the whole day, a school 

36 



EXERCISES OF PRACTICAL LIFE 

lunch is necessary, the serving of which is 
largely the work of the children. 

No feature of the Montessori schools has been 
more commented upon than the skill and deft- 
ness with which the children serve these lunch- 
eons. Every one who has been present at one 
of these luncheons will recall with pleasure the 
eager yet serious interest exhibited by the chil- 
dren, and the success with which tiny tots did 
what we usually associate solely with older 
hands. It seems to the writer that we have here 
once more an instance of putting the school 
exercises on the plane of normal child-life. Not a 
few kindergartners — perhaps most of them — 
know from their own experience how much 
pleasure children take in such real life matters. 
The interest is just what we have a right to 
expect. But what about the skill? It has been 
said that "not a mistake is made, not a glass is 
broken, not a drop of soup is spilled." And many 
friends of the system have asserted that this suc- 
cess is due to the muscular control gained from 
the use of the didactic apparatus. To these 
assertions, two remarks may be made. First, 
although the children do exceedingly well, they 
still do make lapses. The writer saw soup spilled 
and mistakes made in distributing lunch-baskets. 

37 



THE MONTESSORI SYSTEM 

His friend saw a glass and a plate broken. Sec- 
ond, there were no evidences of greater skill or 
ability than could reasonably be expected from 
the amount of attention paid by the school to 
the specific exercise itself. Here, as elsewhere, 
"practice makes perfect." 

The question that concerns us, however, fc, 
rather the value of such "practical life" activi- 
ties to our American schools. The long school 
day is well worth consideration. Where mothers 
are so closely confined to duties either at home or 
on the outside that the children cannot receive 
proper attention, all-day care of the young chil- 
dren by the kindergarten would be highly desir- 
able. Again, in large cities, where opportunities 
for open air play are few or difficult of manage- 
ment, all children alike would probably benefit 
from regularly supervised playground exercises. 
If, then, the kindergartens for the very poor 
everywhere, and for practically all classes in the 
large cities, could have an all-day session with 
much time spent in the open air, the results 
would probably be highly beneficial. The admin- 
istrative difficulties connected with such enlarged 
functions of the kindergarten, while great, would 
not prove insuperable, if only the desirability of 
such changes were admitted. 

38 



EXERCISES OF PRACTICAL LIFE 

What " practical life" exercises should be in- 
cluded in any particular kindergarten will clearly 
depend upon the community. If the mothers of 
the children who attend the kindergarten are too 
busy, or too careless, to see that the children 
appear in the morning clean and neatly dressed, 
evidently, the duty of that kindergarten is differ- 
ent from what it would be if the children came 
from another class of homes. This is to say 
nothing more than that the curriculum of any 
school should be a reflex of the needs of the 
locality served — a principle well recognized in 
present-day school theory. 

In this connection we give most cordial ap- 
proval to the Montessori practice of letting the 
child do for himself as far as this may be feasible. 
In the homes of the poor, necessity may force the 
child to attend to his wants. Within limits, no 
training could be better. Among the wealthier, 
the over-zealous nurse or the indulgent mother 
too often strives to anticipate each want and 
effort of the child. No service could be worse 
directed. As we have already discussed, the 
claims of morality and intelligence alike demand 
that thd[ individual come into first-hand contact 
with actual situations of thwarted impulse.\ The 
personal striving and contriving incident to meet- 

39 



THE MONTESSORI SYSTEM 

ing such situations are most wholesome, both to 
forming intelligent self-reliance and to furnishing 
the organized data necessary for meeting other 
situations. Here again we hope much from the 
Montessori corroboration of a doctrine long and 
widely preached, but too often disregarded. 

The general idea of including among the school 
exercises such occupations as are mainly valu- 
able from demands of immediate utility is one 
that proves attractive. It is well recognized that 
cooking as a school subject, for example, does 
not arouse the same serious interest among our 
pupils that it formerly aroused in the home, when 
the girl who took it up did so to meet the immedi- 
ate need in the household. The motivation, as 
we say, is largely lacking in the artificial situa- 
tion of the schoolroom. If, now, the school can 
bring into its service something of the gripping 
interest that attaches to actual and immediate 
social demand, we shall have the real effort that 
counts. It must be admitted, however, that this 
will not hold of all the " practical life" activities, 
because some of the most insistent of these have 
never aroused in young children any great inter- 
nal motivation even in the best homes: washing 
the face and hands, for example. In such cases, 
the social approval or disapproval of the school- 
40 



EXERCISES OF PRACTICAL LIFE 

room may prove distinctly helpful in fixing a 
habit that might never be learned in inferior 
home surroundings. 

While no one could suppose that a curriculum 
devised for a particular class in Rome would 
serve, unmodified, in America, we have no hesi- 
tation in concluding that we can find suggestion 
for thought in the long school day, in the prac- 
tical effort to adapt the school exercises to the 
needs of the community, and in the possible 
increase of motivation by the introduction of 
activities the demand for which is immediate and 
actual. The whole conception is but part of the 
world-wide demand that the school shall func- 
tion more definitely as a social institution, adapt- 
ing itself to its own environment and utilizing 
more fully actual life situations. 



VII 

SENSE-TRAINING BY MEANS OF THE 
DIDACTIC APPARATUS 

No topic is more fundamental to the Montessori 
method, as understood by its author and her fol- 
lowers, than is its system of sense-training. This 
was Madam Montessori's initial approach to the 
study of education; and throughout, it has deter- 
mined her general point of view. One fourth of 
the exposition of the system as found in her book 
is given to this one topic. The didactic appara- 
tus — the most striking feature of the system to 
the popular mind — was devised to make possi- 
ble a proper training of the senses. Evidently a 
careful consideration of this most fundamental 
part of the system is necessary to any first-hand 
appreciation of the Montessori method. 

While writers on sense - training have not 
always been careful to differentiate their several 
theories, we can easily distinguish three separate 
lines of thinking in this field. The first is that the 
sense-organ itself can be improved. That is, by 
systematic training we can, for example, make 
42 



SENSE-TRAINING 

the eye, as an optical instrument, see more and 
better. One would thus look out of the trained 
eye as through an improved telescope. To this 
notion the two remaining groups of theorists 
unite in entering a protest. It is not the organ 
itself, these say, that is changed; a new brain 
connection has been set up; nothing more. A 
certain color means that this peach is ready to 
eat. The child thereafter looks for that color, 
and notes it when present. A connection has 
been made between a color — present but pre- 
viously unnoticed — and the pleasurable expec- 
tation of eating the peach. In the sense that he 
notices more, the child may be said to see more. 
The difference, however, is not that the optical 
image has been changed; but only that certain 
portions of that image are now differently con- 
nected in the child's conscious world. 

Which of these theories is true? Consider a 
typical case. Contrast Fenimore Cooper's Indian 
with a student of languages. Is there any doubt, 
we may imagine some one asking, that the Indian 
has a keener eye, that he can see more, and more 
distinctly? If the trial of strength be in the forest, 
certainly; the scholar is hopelessly inferior. But 
bring the Indian into a library. Place before him 
a page of Latin and a page of French. The two 

43 



THE MONTESSORI SYSTEM 

will appear to him alike, a blur of little marks. 
One glance, however, tells the scholar that the 
one is Latin and the other French. Which eye, 
then, really sees the better? Is it not clear that 
in this case each one sees according to the experi- 
ence he has had, according to the connections 
that have been set up? We may safely accept the 
judgment of the authorities that the sense- 
organs of the normal child, after the age of two 
or three, do not in themselves change by train- 
ing. 

Let us now differentiate the second and third 
theories. These agree in saying that sense- 
training is a matter of making brain connections. 
They differ as to how specialized the effect of 
such training is. The second of the three theories 
says that if the child has learned to discriminate 
a certain group of visual forms, he has trained his 
power of visual discrimination so that thereafter 
he can the better discriminate any matter of 
sight. The third says that there is no general 
power or faculty of visual discrimination or of 
anything else, that training along one line will 
carry over into another line only in the degree 
that the two lines of activity have common 
elements. The discussion of this point is too 
lengthy to enter upon here. It suffices to say 

44 



SENSE-TRAINING 

that after a great deal of investigation there is 
substantial agreement on the general statement 
of the third theory as made above. Although 
there yet remain differences of opinion as to what 
constitute common elements, the old notion of 
the existence of faculties of the mind and their 
consequent general training is now entirely 
rejected by competent psychologists. We no 
longer speak of judgment as a general power that 
can be trained; nor of discrimination, nor of 
observation. 

What practical difference would it make which 
theory any one might hold? Pedagogically, the 
applications of the several theories will be widely 
divergent. If one hold to either the first or the 
second theories, he will feel that training of some 
gymnastic kind is the desideratum. He will say, 
refine the sense of sight; train the general power 
of visual discrimination. And so, generally, no 
matter what specific activity you engage in, it 
is the training that counts. The sense-qualities 
of objects which you may happen to learn are 
of relative insignificance; it is the refinement of 
sense that we seek. The third theory says, on the 
contrary, it is what you learn that counts. If it 
be a matter of sense-training, then learn to make 
those discriminations of color, form, or other 

45 



THE MONTESSORI SYSTEM 

sense-quality, that will enter fruitfully into your 
subsequent life. 

Where now stands Madam Montessori? In 
general she puts great emphasis upon "the edu- 
cation of the senses." That she accepts either 
the first or second theory would seem to be justi- 
fied by such a statement as "we must not con- 
fuse the education of the senses with the concrete 
ideas which may be gathered from our environ- 
ment by means of the senses." And similarly, 
when she speaks of blindfolding the child "for 
the education of the senses in general, such as 
in the tactile, thermic, baric, and stereognostic 
exercises." Accordingly, such phrases as "edu- 
cation of the stereognostic sense," "education of 
the chromatic sense," suffice to show either care- 
lessness in thinking or erroneous theory. Sim- 
ilarly for such statements as "the education of 
the senses makes men observers," "before he 
can become a doctor, he must gain a capacity 
for discriminating between sense-stimuli." These 
statements certainly seem to imply a belief in the 
validity of the general transfer of training; and 
the more one studies Madam Montessori's writ- 
ings, the more convinced does he become that 
she holds to some such position. Apparently she 
vacillates between the first and second theories. 
46 



SENSE-TRAINING 

Perhaps the most unmistakable assertion of her 
position is the following: "It is exactly in the 
repetition of the exercises that the education of 
the senses consist; their aim is not that the child 
shall know colors, forms, and the different quali- 
ties of objects, but that he refine his senses 
through an exercise of attention, of comparison, 
of judgment. These exercises are true intellectual 
gymnastics. Such gymnastics, reasonably di- 
rected by means of various devices, aid in the 
formation of the intellect, just as physical exer- 
cises fortify the general health and quicken the 
growth of the body." l 

Here we have most of the earmarks of the old 
theory of general discipline: "not that the child 
shall know colors, forms . . . but that he refine his 
senses " ; " intellectual gymnastics " ; and the same 
old analogy of mind and body. After the writer 
had read Madam Montessori's book and had 
studied the apparatus, he was anxious to ascer- 
tain at first hand, if he could, her opinion on the 
question of the general transfer of mental train- 
ing. The interview was difficult, as the inter- 
preter was not versed in psychology; but the 
writer came away convinced that Madam Mon- 
tessori had up to that time not so much as heard 
1 The Montessori Method, p. 360. 

47 



THE MONTESSORI SYSTEM 

of the controversy on general transfer; and that 
she still held to the doctrine of formal discipline 
discarded years previously in both Germany and 
America. 

Important as it is to establish Madam Mon- 
tessori's intention in devising the didactic appa- 
ratus, it is equally important — perhaps more so 
— to ask what is the actual effect of the appara- 
tus when used by children. If we set aside the 
various buttoning and fastening apparatus as 
belonging, at least indirectly, to the " practical 
life" activities, and certain other pieces of 
apparatus to be discussed in connection with 
writing, all that is left calls for some form of 
sense-discrimination. This apparatus is so de- 
vised that discriminations may be made either 
between widely varying stimuli, as a ten-centi- 
meter cube and a one-centimeter cube, or be- 
tween those which differ only slightly, as the 
nine-centimeter cube and the eight-centimeter 
cube. It is a matter of common as well as of 
scientific knowledge that practice with a series 
of graduated stimuli results in finer discrimina- 
tions made more quickly than was at first possi- 
ble. This will be true, in all probability, of any 
normal child who may deal with the Montessori 
apparatus. He will, for example, learn to discrim- 

4 8 



SENSE-TRAINING 

inate infallibly, perhaps, and almost instantly 
the several weight blocks. If one please so to 
term it, he has educated his "baric sense." But 
does this mean that he will be able to distinguish 
weights in all the ordinary affairs of life? By no 
means. It may be that he will never use his 
specific skill at all. He will surely never use it 
directly, unless there should chance to come a 
demand for the discrimination of small weights 
under conditions quite similar to those under 
which the skill was acquired. His skill, for ex- 
ample, would not suffice to tell whether any given 
letter would go for two cents. The formal train- 
ing with the weight blocks would not prove a 
sufficient substitute for practice with the weight 
of letters. Has he, then, gained absolutely noth- 
ing that will carry over? The extent to which he 
has profited is still under dispute. He has cer- 
tainly added to his concept of weight. But the 
value of this increment is a comparative one, 
depending upon what he already had got or 
would otherwise incidentally get, and on the 
present need for the concept. In general, it 
seems true that the really necessary concepts, 
such as of hardness, of heat, or of weight, etc., 
come in the normally rich experience of the 



49 



THE MONTESSORI SYSTEM 

child-life; and conversely, those that do not so 
come are not then necessary. 

In the same way, the training got with each of 
the several pieces of the didactic apparatus is 
genuine, but highly specialized, and along lines 
for the most part so removed from ordinary life 
conditions that the probability of its function- 
ing directly as skill is very remote: too remote, 
except possibly with the colors, for us to desire 
the particular skill attained. Of what advantage 
will it be to the child to recognize that this given 
cylinder fits into the second hole of the doubly 
varying cylinder box? If one should reply that 
the advantage lies in bringing the child to make 
discriminations, we should again be in the midst 
of the discussion as to the general transfer of 
training. The specific result is certainly the 
training in this particular discrimination, and 
not a general power. 

It is true that any experience with color or 
form or weight helps to make one's concepts of 
these things; and pleasurable experience along 
any one of these lines will lead the child to look 
for further allied experiences. It is further true 
that growth comes from the organization of such 
experiences, and that this is the training that we 
really wish. In these ways exercise with this 
SO 



SENSE-TRAINING 

apparatus may be, indeed, will be, of service; 
because from it comes opportunity for the con- 
scious consideration of such experiences. The 
formal and mechanical aspect of the training 
is, however, practically valueless. Any play in 
which the consideration of a size-experience, for 
example, enters, will do just as well as does the 
broad stair of the didactic apparatus. All must 
approve Madam Montessori's wish to provide 
more fruitful sense-experiences. Most children 
need more activity of this kind. The natural 
fondness of the young child for manipulation and 
the like is sufficient proof of the fact. Care, too, 
is necessary that the opportunities offered be 
sufficiently varied and sufficiently ordered to 
bring the desired richness of experience. But 
these considerations — all important though 
they be — afford no support for the dogma of 
general transfer, nor do they call for an apparatus 
so formal and mechanical in character as the 
system under review offers. 

We must, then, take exactly the opposite view 
from Madam Montessori as to the nature of 
sense- training. She says that the "aim is not 
that the child shall know colors, forms, and the 
different qualities of objects." We say that the 
aim is exactly that he may know such things, 

51 



THE MONTESSORI SYSTEM 

and we don't care about his getting any sense- 
training outside of this. We conclude, accord- 
ingly, that Madam Montessori's doctrine of 
sense-training is based on an outworn and cast- | 
off psychological theory ; that the didactic appa- 
ratus devised to carry this theory into effect is in 
so far worthless; that what little value remains 
to the apparatus could be better got from the 
sense-experiences incidental to properly directed 
play with wisely chosen, but less expensive and 
more childlike, playthings. 



VIII 

THE SCHOOL ARTS: READING, WRITING, AND 
ARITHMETIC 

No small interest has attached to the reported 
ease with which children of the Montessori 
schools learn to read and write. In the popular 
mind, this comes in an almost occult manner 
from that individual development which has re- 
sulted from the sense-training of the didactic 
apparatus. A system in which such tangible 
results ensue from so tangible a set of apparatus 
is bound to attract attention. When, moreover, 
it is reported that the children in these marvelous 
schools are left entirely free, and, as it were, play 
themselves into this learning, the acme of educa- 
tional achievement seems, indeed, at hand. 

One acquainted, however, with the history of 
education is prepared to hear of remarkable suc- 
cesses attending the enthusiasm of a new project. 
Pestalozzi's visitors gave accounts of his success 
that seem wonderful even to-day. The monitorial 
schools were similarly acclaimed as ushering in a 

S3 



THE MONTESSORI SYSTEM 

new era of ease and rapidity in learning. Base- 
dow's daughter, Emilie, apparently surpassed 
all in her marvelous acquisition of new languages. 
Ten weeks at the age of three and a half gave her 
French, and the next year an equal time gave 
both a speaking and a reading knowledge of 
Latin. Enthusiasm and the exceptional case 
always account for much. If there be a perma-* 
nent contribution, it must be more tangible. 
Proper scrutiny must be able to find it. 

Madam Montessori teaches the beginnings of 
reading and writing simultaneously. For easier 
criticism we shall treat the two separately. When 
we examine the accounts of the teaching of read- 
ing in the Montessori schools we find an intelli- 
gent utilization of the phonetic character of the 
Italian language. In this language, to speak 
generally, one sign represents one sound, and 
vice versa. The method of teaching, then, is to 
associate the sounds (but not the names) of the 
several letters with their forms, beginning with 
the vowels. The names of the letters are not used 
during this early period. With a one-to-one 
correspondence of sounds and symbols, the whole 
alphabet can be readily taught. It is thus easy 
to build up with letters any given word, or, con- 
versely, to call any word by recalling the sounds 
54 



THE SCHOOL ARTS 

attached to its several letters. If, however, one 
should try to apply this method in America, the 
unphonetic character of the English language 
would present formidable difficulties. Any at- 
tempt to meet these difficulties could but result 
in a plan identical with one or another of the 
quasi -phonetic methods familiar enough to 
American primary teachers. It thus turns out 
that the Montessori method of teaching reading 
has nothing of novelty in it for America. What 
it can offer has long been present with us, and a 
vogue previously won has for a decade been pass- 
ing away. 

When we come to writing, the question is 
somewhat different. Here a special technique has 
been worked out. From some observations on 
an indirect method of teaching a defective to sew, 
Madam Montessori "saw that the necessary 
movements of the hand had been prepared without 
having the child sew." From this she concluded, 
in relation to any complex activity, that "we 
should really find the way to teach the child how, 
before making him execute the task" "Prepara- 
tory movements could be carried on and reduced 
to a mechanism, by means of repeated exercises 
not in the work itself, but in that which prepares 
for it. Pupils could then come to the real work, 

SS 



THE MONTESSORI SYSTEM 

able to perform it without ever having directly 
set their hands to it before." 

Following up this idea, the process of writing 
was analyzed into two essential elements, "the 
muscular mechanism necessary in holding and 
managing the instrument of writing" and "the 
visual-muscular image of the alphabetical signs." 
Special exercises were devised to give the child 
simultaneous training in these two elements of 
writing. The apparatus for the first consists of 
metal geometrical figures and colored pencils. 
The child takes a triangle, for example, draws 
about it, and then with a crayon fills in the figure 
so made. With practice of this sort he gains con- 
trol in the use of the pencil. The apparatus for 
giving the second element consists of sand-paper 
letters mounted on cards and a box of alphabets 
cut from paper. These letters of both kinds are 
in clear script. It is in connection with this sec- 
ond element that the reading is taught. When the 
association of the sound with the form is being 
taught, the sand-paper letters are used; and the 
child is required to trace each letter with his 
index finger as if writing it. They are encour- 
aged to do this repeatedly even with the eyes 
shut. The child is thus gaining at the same time 
both visual and muscular images of the letter and 
56 



THE SCHOOL ARTS 

associating these with each other and with the 
sound. As soon as the child knows some of the 
vowels and consonants, the box of letters is put 
before him. The directress selects a simple word, 
pronounces it so clearly as to analyze it into its 
constituent sounds, and calls for the correspond- 
ing letters. The next exercise is for the child 
to read a word set before him. This he does by 
calling in succession the sounds corresponding to 
the several letters. It is evident how essential is 
the phonetic alphabet to the success of the plan. 
After both the elements of the writing process, 
carried on thus simultaneously, are well fixed, it 
is a simple matter to have the child write. In- 
deed, according to reports, this takes place so 
suddenly as to warrant the phrase "exploding 
into writing. ,, It is easy to believe this, because 
the manipulation of the pencil, the muscular 
image of the word, and the knowledge of the 
value of the letters are all present. The second 
and third of these having already been joined, it 
only remains to connect the first with these two. 
This is the more readily done since the gradual 
perfection of the first and second, even as sepa- 
rate activities, has all the while been bringing 
them closer together. The result is writing. It 
only remains to be said that this writing, while 
57 



THE MONTESSORI SYSTEM 

very slow, is unusually good. The beauty of the 
writing, quite as much as the reported ease of 
acquisition, has brought the system into favor- 
able publicity. 

The appraisal of Madam Montessori's contri- 
bution in the case of writing is difficult. On the 
whole, it appears probable that she has in fact 
made a contribution. Of how much value this 
can prove to those who use the English language 
is uncertain. Probably experimentation only 
can decide. Her plan seems so dependent on 
some single-letter method of learning to read that 
many will be unwilling even to try it, feeling that 
previous experimentation on this point is con- 
clusive. The suggested analysis of the process 
might, however, prove helpful, even if not used 
exactly as she proposes. We shall await with 
interest the results of further discussion and 
experimentation in this field. 

As to arithmetic, there is little to be said. 
About the only novelty is the use of the so-called 
long stair. This consists of ten blocks, of lengths 
varying from one to ten decimeters, being in 
other dimensions the same. These are divided 
into decimeters, alternate divisions being painted 
in like colors. These blocks are used in teaching 
the various combinations which sum ten. On 

58 



THE SCHOOL ARTS 

the whole, the arithmetic work seemed good, but 
not remarkable; probably not equal to the better 
work done in this country. In particular there is 
very slight effort to connect arithmetic with the 
immediate life of the child. Certainly, in the 
teaching of this subject, there is for us no funda- 
mental suggestion. 

The question of introducing the "three R's" 
into the kindergarten period demands separate 
consideration, although it is in part bound up 
with the question of the difficulty of teaching 
them. In this country we seem pretty well agreed 
that these subjects had, as a rule, better not be 
taught prior to the age of six. There is, however, 
no definite experimental basis for such a judg- 
ment; and from this point of view it may well be 
claimed that the question is as yet an open one. 
Education, however, is much more than the ac- 
quisition of knowledge from books. And there 
is reason to fear that the presence of books makes 
more difficult that other part of education. If 
there be any truth in this point of view, it would 
seem to hold particularly of the earlier stages of 
education. From this standpoint some would 
feel that reading and writing might better be 
postponed to a later period than put forward to 
an earlier one. This would not mean that to 

59 



THE MONTESSORI SYSTEM 

learn to read and write is in itself very difficult 
for a child of six; often the reverse is true; but 
that the presence of these tends to divert the 
attention of parent, teacher, and child from 
other and, for the time, possibly more valuable 
parts of education. Education is life; it must 
presume first-hand contact with real vital situa- 
tions. The danger in the early use of books is 
that they lead so easily to the monopoly of set 
tasks foreign to child nature, lead so almost 
inevitably to artificial situations devoid alike of 
interest and vital contact. An unthinking public 
mistakes the sign for the reality, and demands 
formulation where it should ask experience, 
demands the book where it should ask life. The 
writer agrees, therefore, with those who would 
still exclude these formal school arts from the 
kindergarten period. To him a school for the 
young without books is FroebePs chiefest glory. 



IX 

CONCLUSIONS 

We have passed in review the principal features 
of the Montessori theory and practice. Good 
points and bad have appeared. Before attempt- 
ing a summation of the several valuations made, 
it may be well to ask, Where among other sys- 
tems of education does this one belong? What is 
the relation of Madam Montessori to the world's 
educational thinkers? 

When the surmise was made in the first chap- 
ter that Pestalozzi formed the background of 
Madam Montessori's educational philosophy, 
one might better have said that it was the 
Rousseau-Pestalozzi-Froebel group which formed 
that background, although there are more dis- 
tinct marks of Pestalozzianism than of the others. 
This group of educational thinkers are differenti- 
ated from others by the presence of several char- 
acteristics which we find also in the Montessori 
theory. The revolutionary attitude, the feeling 
that one is breaking with customary practice, 
while certainly present, need hardly be men- 
61 



THE MONTESSORI SYSTEM 

tioned, as this is an element found to a greater or 
less degree in all reformers. More to the point 
are: (i) a belief that the child nature is essen- 
tially good; (2) that the educational process is 
fundamentally an unfolding of what was given 
at birth; (3) a consequent belief in liberty as the 
necessary condition of this development; (4) the 
utilization of sense -experiences as means to 
bringing about the development; (5) a tendency 
to accept the faculty psychology; (6) the conse- 
quent tendency to emphasize the disciplinary 
aspect of sense- training; and finally (7) the em- 
phasis upon nomenclature in connection with 
sense-experiences. While not all of these are 
found with distinctness in the writings of each 
one of the group, they either are so present or 
have been drawn as corollaries by followers. 
They are likewise present in the Montessori the- 
ory. When we consider that each of these char- 
acteristic doctrines, while containing a greater 
or less amount of truth, still has needed to be 
strictly revised in order to square with present 
conceptions; when we further consider that 
Madam Montessori's own conception of these 
doctrines has needed an almost identical revision; 
when we still further consider that Madam 
Montessori has confessedly been most influenced 
62 



CONCLUSIONS 

by Seguin, whose ideas were first published in 
1846; when we consider, in particular, that 
Madam Montessori still holds to the discarded 
doctrine of formal or general discipline, — in the 
light of all these, we feel compelled to say that 
in the content of her doctrine, she belongs essen- 
tially to the mid-nineteenth century, some fifty 
years behind the present development of educa- 
tional theory. 

If we compare the work of Madam Montessori 
with that of such a writer and thinker as Pro- 
fessor Dewey, we are able to get an estimate of 
her worth from still a different point of view. 
The two have many things in common. Both 
have organized experimental schools; both have 
emphasized the freedom, self-activity, and self- 
education of the child; both have made large use 
of "practical life" activities. In a word, the two 
are cooperative tendencies in opposing intrenched 
traditionalism. There are, however, wide differ- 
ences. For the earliest education, Madam Mon- 
tessori provides a set of mechanically simple 
devices. These in large measure do the teaching. 
A simple procedure embodied in definite, tangible 
apparatus is a powerful incentive to popular 
interest. Professor Dewey could not secure the 
education which he sought in so simple a fashion. 

63 



\ 



THE MONTESSORI SYSTEM 

Madam Montessori was able to do so only be- 
cause she had a much narrower conception of 
education, and because she could hold to an 
untenable theory as to the value of formal and 
systematic sense-training. Madam Montessori 
centered much of her effort upon devising more 
satisfactory methods of teaching reading and 
writing, utilizing thereto in masterly fashion the 
phonetic character of the Italian language. 
Professor Dewey, while recognizing the duty of 
the school to teach these arts, feels that early 
emphasis should rather be placed upon activities 
more vital to child-life which should at the same 
time lead toward the mastery of our complex 
social environment. Madam Montessori, in a 
measure following Pestalozzi, constantly uses 
logically simple units as if they were also the 
units of psychological experience. In reading and 
writing, it is the letter and the single sound, not 
the word or thought connection, that receive 
attention. Sense-qualities are taught prefer- 
ably in isolation, apart from life situations. She 
speaks also of leading the child "from sensations 
to ideas . . . and to the association of ideas." 
Professor Dewey insists that the experience is the 
unit, and that the logically simple units emerge 
for consciousness by differentiation from the 

6 4 



CONCLUSIONS 

experience. Things, as a rule, are best taught, 
then, in connection with what is for the child a 
real experience, when they enter as significant 
parts into such an experience; and this because 
learning is essentially the differentiation and 
organization of meanings. It is, of course, to be 
borne in mind that a child experience is vastly 
different from the adult experience. What to a 
child is a whole satisfying experience, to us may 
be very fragmentary and disconnected. 

But there are even more comprehensive con- 
trasts. Madam Montessori hoped to remake 
pedagogy; but her idea of pedagogy is much 
narrower than is Professor Dewey's idea of ed- 
ucation. His conception of the nature of the 
thinking process, together with his doctrines of 
interest and of education as life, — not simply a 
preparation for life, — include all that is valid 
in Madam Montessori's doctrines of liberty and 
sense-training, afford the criteria for correcting 
her errors, and besides, go vastly farther in the 
construction of educational method. In addition 
to this, he attacked the equally fundamental 
problem of the nature of the curriculum, saw it 
as the ideal reconstruction of the race achieve- 
ment, and made substantial progress toward a 
methodology of its appropriation. This great 

65 



THE MONTESSORI SYSTEM 

problem of the curriculum, it can almost be said, 
Madam Montessori has, so far, not even seen. 
While this is no adequate recital of Professor 
Dewey's contributions, it suffices, in connection 
with what has been previously said, to show that 
they are ill advised who put Madam Montessori 
among the significant contributors to educational 
theory. Stimulating she is; a contributor to our 
theory, hardly, if at all. 

Is this, then, the final judgment of Madam 
Montessori's contribution? The question of a 
permanent contribution turns on whether there 
have been presented original points of view ca- 
pable of guiding fruitfully educational procedure. 
What novel and original ideas have we found 
that could at the same time bear the scrutiny of 
criticism? The scientific conception of education 
is certainly valid. Madam Montessori may, in a 
way, have come upon it herself; but no one could 
say that the world did not have a fuller concep- 
tion of it prior to her. The most that can be 
claimed on this point is that her advocacy and 
example have proved stimulating. Her doctrine 
of education as unfolding is neither novel nor 
correct. In the doctrine of liberty she has made 
no theoretical contribution; though probably her 
practice will prove distinctly valuable. Our 
66 



CONCLUSIONS 

kindergartens and primary schools must take 
account of her achievement in this respect. Her 
doctrine of auto-education will at most provoke 
thought; the term is good, the idea old. Her 
utilization of " practical life" activities, more 
specifically her solution of early tenement-house 
education, must prove distinctly suggestive. It 
may well turn out that the Casa dei Bambini is 
after all her greatest contribution. The sense- 
training which to her seems most worth while, 
we decline to accept except in a very modified 
degree. The didactic apparatus we reject in like 
degree. Her preparation for the school arts 
should prove very helpful in Italy. It is possible 
that her technique of writing will prove useful 
everywhere. If so, that is a contribution. With 
this the list closes. We owe no" large point of view 
to Madam Montessori. Distinguishing contribu- 
tion from service, she is most a contributor in 
making the Casa dei Bambini. Her greatest serv- 
ice lies probably in the emphasis on the scientific 
conception of education, and in the practical 
utilization of liberty. 



OUTLINE 

I. INTRODUCTION 

i. The genesis of Madam Montessori's educational 

ideas i 

2. Her indebtedness to Seguin 2 

3. Influence of science upon her educational system 3 

4. Insufficiency of her science 3 

5. The Gasa dei Bambini a scientific creation . . 5 

II. EDUCATION AS DEVELOPMENT 

1. Education as development is frequently mis- 
conceived 7 

2. Individual development is possible only through 
the utilization of the race achievement .... 8 

3. Madam Montessori considers development to be 
the mere unfolding of latency 9 

4. Such a conception (a) but ill adapts to a changing 
environment, and (b) furnishes a wrong basis for 
freedom 10 

III. THE DOCTRINE OF LIBERTY 

1. The doctrine of liberty dates from Rousseau . 12 

2. Madam Montessori wrongly bases liberty upon 
the unfolding of latency . . , 12 

68 



OUTLINE 

3. Child liberty as the pre-condition of child study 13 

4. The Montessori doctrine of liberty shown by 
contrast with the conservative kindergarten . . 14 

5. Montessori liberty not anarchy 14 

6. The advantages of child liberty 17 

7. Cooperation on the basis of liberty 19 

8. Amount of knowledge and skill expected of a 
child of six 21 

9. Such knowledge and skill best secured on a basis 

of freedom 22 

10. Wrong conduct best avoided by encouragement 

of proper impulses 24 

11. The child best learns to "get on" with others in 

a regime of freedom 24 

12. Madam Montessori^ reemphasis of freedom 
highly commendable 26 



IV. ADEQUACY OF SELF-EXPRESSION IN THE 
MONTESSORI SYSTEM 

1. Self-expression is the essence of freedom ... 27 

2. The didactic apparatus affords but meager self- 
expression, 27 

3. Imaginative and constructive play has little or 

no place in the Montessori schools 28 

4. Games, drawing, modeling, stories, and drama- 
tization are inadequately utilized ..... 28 

5. The "practical life" activities afford excellent 
opportunities of self-expression 29 

69 



OUTLINE 

6. On the whole, the Montessori curriculum is 
unduly restrictive 30 

V. AUTO-EDUCATION 

1. Some scheme of auto-education is the necessary 
counterpart of a regime of freedom .... 31 

2. Madam Montessori seeks auto-education through 
mechanically simple didactic apparatus . . . 33 

3. Such auto-education is too restricted .... 33 

4. Self -education comes best where real problems 
present themselves naturally 34 

VI. EXERCISES OF PRACTICAL LIFE 

1. The tenement-house origin of the Casa dei 
Bambini introduced many immediately practical 
features 36 

2. Many of these exercises delight the children and 
give excellent training 37 

3. The longer school day is worthy of consideration 

for America 38 

4. Specific practical exercises must fit community 
needs 39 

5. Children should be encouraged to do for them- 
selves 39 

6. The motivation of actual service 40 

7. Madam Montessori's "practical life" exercises 
are in keeping with a world-wide educational 

>• trend 41 

70 



OUTLINE 

VII. SENSE -TRAINING BY MEANS OF THE 
DIDACTIC APPARATUS 

i. Sense-training is most fundamental in the Mon- 

tessori system 42 

2. Improving the sense-organs vs. brain connections 42 

3. General vs. specific training 44 

4. Pedagogical corollaries of the three suggested 
theories 45 

5. Madam Montessori accepts the rejected theories 46 

6. The didactic apparatus is thus based on error . 52 

7. Less formal apparatus would give more useful 
sense experiences 52 

VIII. THE SCHOOL ARTS: READING, WRITING, 
AND ARITHMETIC 

1. The enthusiasm of new educational projects often 
shows remarkable results 53 

2. The Montessori reading method depends on the 
phonetic character of the Italian language . . 54 

3. The two elements of writing are taught sepa- 
rately: 

a. Manipulation of the instrument .... 56 

b. The visual-muscular image of the letters . 56 

4. Whether the Montessori writing method is a 
contribution is as yet undecided 58 

5. The arithmetic teaching has little or no sugges- 
tion for America 58 

6. The "three R's" should probably not be taught 
before the age of six ..,..,... . 59 

71 



OUTLINE 

IX. CONCLUSIONS 

i. Madam Montessori belongs to the Rousseau- 
Pestalozzi-Froebel group of educators . . . . 61 

2. Her system falls essentially below the best 
American theory 63 

3. Her greatest original contribution is the Casa 

dei Bambini as a social institution 67 

4. Her greatest service lies in her emphasis upon a 
scientific education and in the practical utiliza- 
tion of liberty 67 



RIVERSIDE EDUCATIONAL 
MONOGRAPHS 

GENERAL EDUCATIONAL THEORY 

Dewey's MORAL PRINCIPLES IN EDUCATION 35 

Dewey's INTEREST AND EPFORT IN EDUCATION 60 

Eliot's EDUCATION FOR EFFICIENCY 35 

Eliot's CONCRETE AND PRACTICAL IN MODERN EDUCATION 35 

Emerson's EDUCATION 35 

Eiske's THE MEANING OF INFANCY 35 

Hyde's THE TEACHER'S PHILOSOPHY 85 

Palmer's THE IDEAL TEACHER 35 

Palmer's TRADES AND PROFESSIONS 35 

Prosser's THE TEACHER AND OLD AGE 60 

Terman's THE TEACHER'S HEALTH 60 

Thorndike's INDIVIDUALITY 35 

ADMINISTRATION AND SUPERVISION OF SCHOOLS 

Betts's NEW IDEALS IN RURAL SCHOOLS 60 

Bloomfield's VOCATIONAL GUIDANCE OF YOUTH 60 

Cabot's VOLUNTEER HELP TO THE SCHOOLS 60 

Cole's INDUSTRIAL EDUCATION IN ELEMENTARY SCHOOLS 35 

Cubberley's CHANGING CONCEPTIONS OF EDUCATION 35 

Cubberley's THE IMPROVEMENT OF RURAL SCHOOLS 35 

Lewis's DEMOCRACY'S HIGH SCHOOL 60 

Perry's STATUS OF THE TEACHER 35 

Snedden's THE PROBLEM OF VOCATIONAL EDUCATION 35 

Trowbridge's THE HOME SCHOOL 60 

Weeks's THE PEOPLE'S SCHOOL 60 

METHODS OF TEACHING 

Bailey's ART EDUCATION 60 

Betts's THE RECITATION 60 

Campagnac's THE TEACHING OF COMPOSITION 35 

Cooley's LANGUAGE TEACHING IN THE GRADES 35 

Earhart's TEACHING CHILDREN TO STUDY 60 

Evans's THE TEACHING OF HIGH SCHOOL MATHEMATICS 35 

Pairchild's THE TEACHING OF POETRY IN THE HIGH SCHOOL 60 

Freeman's THE TEACHING OF HANDWRITING 60 

Haliburton and Smith's TEACHING POETRY IN THE GRADES 60 

Hartwell's THE TEACHING OF HISTORY 35 

Haynes's ECONOMICS IN THE SECONDARY SCHOOL. . 60 

Hill's THE TEACHING OF CIVICS 60 

Jenkins's READING IN THE PRIMARY GRADES 60 

Kilpatrick's THE MONTESSORI SYSTEM EXAMINED . 35 

Palmer's ETHICAL AND MORAL INSTRUCTION IN THE SCHOOLS.. .35 

Palmer's SELF-CULTIVATION IN ENGLISH 35 

Suzzallo's THE TEACHING OF PRIMARY ARITHMETIC 60 

Suzzallo's THE TEACHING OF SPELLING 60 

igi6 





1 (' 



► 21-1C 



14 DAY USE 

RETURN TO DESK FROM WHICH BORROWED 

EDUCATION-PSYCHOLOGY 
LIBRARY 

TEL. NO. 642-4209 

This book is due on the last date stamped below, or 

on the date to which renewed. 

Renewed books are subject to immediate recall. 




MAR 2 7 1974 


— — 


?• 


k . """" k ' 




MAR 2 8 REC'U »g AM 






1979 












RECEIVED 






MAY 1 Q IQflK *liAl! 






1YIMT ly lyoo 1U AIV 






EDUC-PSYCH. LIBRARY 






















• 




















LD 21A-15m-ll,'72 General Library 
(Q5761S10)476 — A-32 University of California 

Berkeley 





Jb J40M 




UNIVERSITY OF CALIFORNIA LIBRARY